Click the Start a free trial link to start a 15-day SaaS trial of our product and join our community as a trial user. If you are an existing customer do not start a free trial.
AppDynamics customers and established members should click the sign in button to authenticate.
Recently we've configured Health rules for set of applications.
We've started to receive alerts for Memory utilization like it breaches the warning threshold value of 75%.
Could someone pls tell me if I receive this kind of alerts how should I start my analysis to pin the problem.
What are the parameters/graphs to lookout for.?
The embedded machine agent which comes with .net agent periodically collects machine snapshots.
When you recieve alerts for high memory usage, check the time frame and machine it is orginating from and go to Machine Snapshots tab on application dashboard. Filter by machine name and time frame and you would get snapshots with high memory usage.
Review few of these snapshots to identify which of the processes taking high memory. After identifying which process is taking lots of memory, indentify if it is an instrumented .net process or any other process running on the server taking high memory.
For .net based instrumented processes -
Check if the process is consistently taking a high memory or if it is increasing with time or if periodically it is going down ? Based on above details you can pinpoint if there is any memory leak issue or not ?
You can also enable object instance tracking for a node to refer top 20 .NET framework classes and the top 20 application classes based upon the number of instances - https://docs.appdynamics.com/display/PRO44/Object+Instance+Tracking+for+.NET
In case if the process is showing symptoms of memory leak, you can take memory dump and you can analyse using Microsoft DebugDiag or Win DBG utility.
Important doc links - https://docs.appdynamics.com/display/PRO44/Monitor+Windows+Hardware+Resources